China
Africa
Explore chapters and articles related to this topic
Fractal Mechanics Is Not Quantum but Original—Geometric
Published in Anirban Bandyopadhyay, Nanobrain, 2020
The entropy of a series of clock outputs from a biological system E=−(lnTmax)−1∑t=1TmaxP(t)lnP(t). The entropy is known as Recurrence period density entropy (RPDE) and used to find the degree of symmetry in a composition of deterministic and stochastic signals (Marwan et al., 2007).
Empirical Bayes Transfer Learning for Uncertainty Characterization in Predicting Parkinson’s Disease Severity
Published in IISE Transactions on Healthcare Systems Engineering, 2018
We present an application of modeling the predictive relationship between Parkinson’s disease (PD) and speech properties from noninvasive speech tests. We download data from Center for Machine Learning and Intelligent Systems at the University of California, Irvine (http://archive.ics.uci.edu/ml/datasets/Parkinsons%2BTelemonitoring) (Tsanas et al. 2010a). There are 42 people with PD diagnosed within the past five years at trial onset, each considered as one domain. Speech signals for those 42 subjects are collected at their home by using a telemonitoring system. The number of signals for each patient varies from 100 to 200 and total number of signal records is 5875. The classical speech signal processing techniques, i.e., dysphonia measures, are applied to all signals for extracting features. All Jitter and shimmer termed features are used to describe the cycle-to-cycle voice variability and amplitude. The harmonics-to-noise ratio and noise-to-harmonics ratio denote the signal-to-noise estimates. The recurrence period density entropy (RPDE) measures the extent of the vocal folds to support simple vibration. The detrended fluctuation analysis (DFA) characterizes turbulent noise in the signal. The pitch period entropy (PPE) addresses the impaired control of stable pitch (Tsanas et al. 2010a). In total, there are 16 dysphonia measures to 5875 sustained phonations for 42 subjects. The typical unified Parkinson’s disease rating scale (UPDRS) is used to reflect the severity of disease symptoms, which is the response variable spanning from 0 to 176.